j. RaptorRes. 31 (3):208-212 ¸ 1997 The Raptor ResearchFoundation, Inc.

BREEDING DENSITY AND LANDSCAPE-LEVEL HABITAT SELECTION OF COMMON BUZZARDS (BUTEO BUTEO) IN A MOUNTAIN AREA (ABRUZZO APENNINES, ITALY)

VINCENZO PENTERIANI Laboratoired•Ecologie, Universitg de Bourgogne,Batiment "Mirande, "B.P. 138, 21004 Dijon Cgdex,France and StazioneRomana Osservazionee ProtezioneUccelli (S.P• O.P..U.), c/o Oasi Naturale W.W.E "Boscodi Palo '• 00055 Ladispoli,Rome, Italy

BRUNO FAIVRE Laboratoired•Ecologie, Universitg de Bourgogne,Batiment "Mirande, "B.P. 138, 21004 Dijon Cgdex,France

ABST•ACT.--Thebreeding density and landscape-levelhabitat selection of Common Buzzards(Buteo buteo)was studiedfrom 1989-93 in a mountain area of Abruzzo Apennines (central Italy). Analysisof landscapefeatures was based on circular plots (2.5 km diameter) centered on occupied nest trees.A total of 32 Common Buzzardnesting territories were identifiedwithin a 387 km2 area (8.3 pairs/100 km2, mean nearest-neighbordistance 2.5 km). The averagealtitude of the nest siteswas 1399 m above sealevel and 73.1% were orientedNE. Stepwisediscriminant function analysis showed significant dif- ferencesbetween nesting (N = 17) and control sites(N = 15) basedon four landscapevariables: relief index, distancefrom forest edge, distancefrom paved road and distancefrom valleybottom. Results suggestthat Common Buzzardsselect nesting areasin the eastern portion of forestswhich are distant from roadsbut closeto valleybottoms, in rugged areasof irregular morphology. K•Y WORDS: CommonBuzzard; habitat selection; landscape-level; breeding density; Buteo buteo.

Densidad de cr/a y selecci0nniveles de paisajehfbitat en Buteobuteo RESUMEN.-•Ladelicadeza de cria y nivel del paisaje,selecci6n del hfbitat del Buteobuteo rue estudiado de 1989-93 en un frea de montafia de Abruzzo Apennines (central Italy). Anflisis del elementosdel paisajeestuvo basado en lugarescirculos (2.5 km difimetro) centrado en nidos de frbol ocupados.Un total de 32 B. bute0territorios de nido fueron identificadosdentro de 387 km2 frea, (8.3 pares/100 km2, media cerca-vecinodistancia 2.5 km). E1altitud regular de los nidosrue 1399 m arribadel mar y 73.1% fueron orientadosNE. Una funci6n discriminantede pasosensefio un anflisiscon diferencias significasentre nidos (N = 17) y sitiosde control (N = 15) basadosen cuatropaisajes variados: relevo indicie, distanciade la orilla del bosque,distancia del camino pavimentadoy distanciadel fondo del valle. ResultadosSugieren que el B. buteoselecciona fireas de nidos en lugares este en el bosquedonde estJn muy lejos de caminospero mas cerca al rondo del valle, en fireastoscas de morfologia irregular. [Traducci6n de Rafil De La Garza,Jr.]

Nest-site preferences have been described for sible effectsof landscapestructure. In this paper, the Common Buzzard (Buteo buteo) (Tubbs 1967, we present a landscape-levelanalysis of Common Glutz von Blotzheim et al. 1971, Tubbs 1974, Arce Buzzardnest sites,which wasconducted to identify Velasco 1987, Taylor et al. 1988), but few studies the landscapedeterminants of nest-siteselection. have attemptedto quantitativelydetermine the fac- METHODS tors involved in nest-siteselection at a landscape A population of Common Buzzardswas studied from level (Newton et al. 1982,Jedrzejewski et al. 1988, 1989-93 in a mountain area of central Italy (AbruzziAp- Kostrzewa 1989, Hubert 1993, Hohmann 1994, ennines). The studycovered a 387 km2 area of beech Graham et al. 1995, Cerasoli and Penteriani 1996). (Fagussylvatica) forest (typicalof the Apennine massifsof the Abruzzi region) that covers the National Park of All of thesestudies have analyzednest-site selection Abruzzi and the Sirente mountains. Elevation of the area at a microhabitat level (nest-tree characteristics rangesfrom 1000-2340 m. The landscapehas a distinct and stand structure) without consideringthe pos- mosaicstructure with large woodland areasand reforest-

2O8 SEPTEMBER1997 LANDSCAPE-LEVELHABITAT SELECTION BY BUZZARDS 209 ed tracts of Pinus nigra, cropland, pasturesand fallow of the analysis.The classificationof the described sites, land from 1000-1800 m elevation. obtainedwith DFA, wastested with Kappa statistic(Titus Occupied nestswere located by systematicfoot search- and al. 1984). The robustness of the nest-site selection es of the area prior to leafout. We alsoused playbacksof model was testedwith a jack-knife procedure. We used a recorded Common Buzzard calls during the months of chi-squaretest to analyzethe selectionof nest-siteslope March-April (prelaying period) and June-July (nesding exposure. and fledgling periods) (Cerasoli & Penteriani 1992). Ar- easwhere a pair of Common Buzzardswas observed dur- RESULTS ing the breeding period, but no nest was found, were classifiedas possiblenesting territories (Jedrzejewskiet Nest-site Density. A total of 26 known and 6 sus- al. 1994). A number of nestingterritories were identified pected Common Buzzard nesting territories were by observingadults carrying nesting material, by noting identified within the 387 km2 studyarea, for a den- where the displaysof males ended with steep divesinto sity of 8.3 pairs/100 km2. Mean distancebetween the woods (Picozzi and Weir 1974) and from alarm calls of adults and shrill calls of the fledged young. nestingterritories averaged 2.5 km (range = 1.62- We used the nearest-neighbordistance method (New- 4.12 km, SD = 0.54). Within woodland areas, Com- ton et al. 1977) to estimatenesting density. Regularity in mon Buzzard nesting siteswere spacedregularly, nest-sitespacing was computed with a G-test (Brown & as shown by the G-test (G = 0.96). Rothery 1978). Landscape-levelanalysis of habitat selec- Landscape-levelHabitat Selection. The average tion only considered those Common Buzzard nest sites altitude of buzzard nest sites was 1399 m above sea where nests had been located. Moreover, all nest sites that changedduring the studyperiod due to road build- level (range 1150-1550 m, SD = 131.87). Analysis ing, cutting of forest tractsor changesin farming were of nest exposure(N = 26) showedthat 73.1% (N excludedfrom the analysis.Analysis of landscapefeatures = 19) were oriented NE (X• = 33.69, df = 3, P = was based on circular plots centered on the occupied 0.001), 3.8% (N = 1) S and SE, and 19.3% (N = nest tree. These plots had a diameter equal to the mean distancebetween neighboring nest sites.Each nest site 5) SW. was characterizedusing a set of 23 variables:slope ex- The DFA showed significant differences (P < posure, elevation, eight variablesdescribing patch com- 0.05) betweennesting (N = 17) and control sites positionof the landscape(percentage of woodlands,pas- (N = 15) based on the four landscapevariables tures, fallow land, fallow land with trees, rocks, crops, cropswith trees and built-up patches),three variablesfor relief index, distance from forest edge, distance horizontal heterogeneity (number of ecotones,number from paved road and distancefrom valley bottom of different habitats calculated on two orthogonal axes (Table 1). We obtained correct classification for 14 from the plot center and patch interspersionindex [hab- of the control sites (93.3%) and 16 of the Common itat changes/plotarea] x 100, calculatedon two orthog- Buzzard nesting sites (94.1%). Conversely,there onal axesfrom the plot center; Baxter and Wolfe 1972), was one misclassified control site (7%) and one two variablesfor vertical heterogeneity(maximum differ- ence in elevation and relief index calculated as the sum misclassifiednesting site (6%). This classificationis of the number of contour lines crossedby two orthogo- 87% better than random (Kappa -- 0.874, Z = nal axes from the plot center;Janes 1985, Litvaitiset al. 4.946, P < 0.0001). The jack-knife classification 1994), and eight variablesfor distanceof nest sitesfrom showed the robustness of the model with 88.2% of surroundinglandscape components (forest opening, for- est edge, valley bottom, built-up area, paved road, path- the nestingsites and 93.3% of the control sitescor- ways,cliffs, permanent water). The number of ecotones, rectly classified. number of habitats and the interspersionrelief indexes were sampledon two straightlines oriented N-S and W-E DISCUSSION along the plot diameters.Areas of each of the different Common Buzzard nesting density decreases habitatswere determined on the basisof land use maps to a scaleof 1:25 000. For each nest site,one control plot from 8.3 pairs/100 kme in the mountain areas of was established where we measured the same variables as the Apennines,to 19.7 pairs/100 kme in the hills in nest site plots, except for slope exposureand elevation in the piedmont,to 32 pairs/100 kme in woodlands to estimate landscape selection. Each control plot was of low-altitude areas (Manzi and Pellegrini 1989, centered around a random point located between nest- site plots. To qualify as control plot, the plot had to lie Manzi et al. 1991). Low nesting densitiesat higher within a forested area. Plots which did not have woodland altitudesis likely due to the scarcityof prey as ev- areas or which had only young plantation areas (where idenced by the lower density of birds in high Common Buzzards do not nest) were not included in our mountain areas (36 pairs/10 ha; Bernoni 1995) analysis(Hubert 1993,Jedrzejewski et al. 1994). than in piedmont (59.2 pairs/10 ha; Pandolfi and Landscapecharacteristics of nest-siteand control plots were comparedby usinga stepwisediscriminant function Taferna 1991) and plain areas (158 pairs/10 ha; analysis(DFA, Sokal and Rohlf 1981). We used the 5% Bernoni et al. 1989). The averagenearest-neighbor level of significancefor including variablesin each step distanceof 2.5 km was also relativelyhigh when 210 PENTERIANI^Nt• FArV• VOL. 31, NO. 3

Table 1. Samplemeans and standarddeviations of landscapehabitat variables measured at control and nestsites of the Common Buzzard.Significant differences determined by StepwiseDiscriminant Function Analysis.

NESTINGSITES (N = 17) CONTROLPLOTS (N = 15) Woodlandpatches (%) 54.5 _+ 22.9 40.2 -+ 25.3 Pasturepatches (%) 24.1 -+ 16.1 19.8 - 11.3 Fallow patches(%) 6.5 + 5 9.4 _ 5.8 Fallowpatches with trees (%) 3.9 _ 4.1 7 --- 7.9 Rocky patches (%) 4.8 _+ 6.5 7.1 -+ 7.7 Croplandpatches (%) 2.7 +- 2.9 3.5 - 5.6 Cropland patcheswith trees (%) 3.5 _+ 8.5 12.4 -+ 14.7 Built-up patches(%) 0 - 0 0.6 -+ 1.3 Number of ecotones 9.8 _ 4.1 16 _+ 4.5 Number of habitats 15.5 _+ 5.6 20 -+ 5.2 Interspersion index 11 +_ 2.4 10.5 -+ 2.1 Maximum difference in elevation (m) 395.1 +_ 165.7 468.3 -+ 121.5 Relief index 47.2 +-- 12.5 23.2 - 9.7* Distancefrom forest opening (m) 267.6 - 155.3 179.3 -+ 140.9 Distance from nearest forest edge (m) 269.1 - 239.22 509.3 -+ 383.4* D•stancefrom valley bottom (m) 983.8 _ 487.7 1438.3 -+ 845.8* D•stance from built-up area (m) 2827.9 -+ 1738.8 2236.7 - 1049.7 D•stance from nearest paved road (m) 1592.6 _+ 1224.4 753.3 - 610.6' Distancefrom footpath (m) 613.2 - 632.1 120 - 88.2 Distance from cliffs (m) 1376.5 +__706.8 1128 - 486.8 Distancefrom permanentwater (m) 1560.3 --_ 959.5 885.3 _+ 443.2 *P• 0.05.

compared with the values of 0.87 and 1.13 km there, both of which are favorite hunting (Newton et al. 1982) 1.04 km (Jedrzejewskiet al. grounds for Common Buzzards. Reliance on 1994) and 1.9 km (Graham et al. 1995) in other open areas for foraging may also explain why areas of Europe. Common Buzzard nest sites are often near forest Our landscapelevel analysisshowed that Com- edges (Tubbs 1974, Knuwer & Loske 1980, Weir mon Buzzards did not select habitat at random at and Picozzi 1983, Goszczynski1985,Jedrzejewski a landscape level, as the majority of nest sites et al. 1988, Kostrzewa 1989, Hubert 1993, Hoh- (94.1%) and control sites (93.3%) were correctly mann 1994, Graham et al. 1995). Open areas classified.These resultssuggest that Common Buz- may also be needed because they facilitate court- zards selectnest sitesin the eastern part of forests ship behavior. Development of higher tempera- that are situatedon northern slopes.The tendency tures and upward thermal air currents over open to use northern slopesmay simply be due to the habitats (Cone 1962, Jedrzejewski et al. 1988, fact that NE facing slopessupport the tallestbeech Cerasoli and Penteriani 1996) may enhance trees, but it may also be related to the fact that courtship flights when pair-bonding takes place these slopesprovide cooler temperaturesand less in the early part of the nesting season.Nest-site sunlight, as well as a denser canopycover that may selection near forest edges may also be attribut- increasenest protection. ed to ease of access to nests and to a need for The Common Buzzard is an area-sensitivespe- an unobstructed view of the surrounding land- cies that requires forested habitatswhich are dis- scape(Roch6 1977, Hubert 1993). tant from roads but close to valley bottoms in ACKNOWLEDGMENTS rugged areas (Robbins et al. 1989). The choice of nest sites which are far from paved roads has We are extremely grateful to Fabio Liberatori, Marina Cerasoli and Francesco Pinchera for their help on the also been corroborated by Kostrzewa (1989). field work;Javier Bustamante,Christine Hubert, Joseph Nesting close to valley bottoms may be due to the K. Schmutz, Camille Ferry and Bernard Frochot gave fact that most pasture and crop lands are found helpful comments on the manuscript.The Administra- SEPTEMBER 1997 LANDSCAPE-LEVELHABITAT SELECTION BY BUZZARDS 211 tion of Abruzzi National Park provided the logistic sup- lation to the abundance of rodents and birds in Bt- port for this work: we thank especiallythe Director, Fran- alowieza National Park, Poland. Ethol. Ecol. & Evol. 6: co Tassi and Cinzia Sulli. 179-190.

LITERATURE CITED KNUW•R,H. ANDICH. LOS•O•.1980. Zur frage des habitat- anspruche des Mausebussards(Buteo buteo) bei der ARCEVELASCO, L.M. 1987. Biologiay ecologiadel Rato- Horstplatzwahl. Vb•elwelt101:18-30. nero Comtn en la regitn Astur-Leonesa.Mercantil- KOSTRZEWA,A. 1989. Nest habitat separation in three Asturias,S.A., Spain. European raptors: Accipitergentills, Buteo buteoand BAXTER,W.L. ANDC.W. WOLFE.1972. The interspersion Pernis apivorus--a multivariate analysis. Pages 553- index asa technique for evaluationof BobwhiteQuail 559 in B.-U. Meyburg and R.D. Chancellor [EDS.], habitat. Proc.Natl. BobwhiteQuail Symp.1:158-165. Raptors in the modern world. Proc. III World Con- BE•tNONI, M., L. IANNIELLO AND P. PLINk 1989. Censi ference on Birds of Prey and Owls. WWGBP/The mento dell'avifauna nidificante in un bosco deciduo Pica Press, Berlin, Germany. dell'Italia centrale. Avocetta 13:25-29. LITV)dTIS,J.A., K. TITUSAND E.M. ANDERSON.1994. Mea- BERNONI,M. 1995. I1 metodo del mappaggio in una suring vertebrate use of terrestrial habitatsand foods. faggeta del Parco Nazionale d'Abruzzo. Avocetta19: Pages254-274 in TA. Bookhout [ED.], Researchand 131. management techniquesfor wildlife and habitats,5th BROWN, D. AND P. ROTItERY. 1978. Randomness and ed. The Wildlife Society,Bethesda, MD U.S.A. local regularity of points in a plane. Biometrica65: MANZl,A. ANDM.R. PELLEGmNI.1989. Dati sulla biologia 115-122. riproduttiva della Poiana (Buteobuteo) in un'area della CERASOLI,M. AND V. PENTERIANI. 1992. Effectiveness of fascia collinare abruzzese. Avocetta 13:109-114. certsusingwoodland birds of prey by playback.Avocetta , M.R. PELLEGRINI, M.S. PELLEGRINI, g. PENTERIANI 16:35-39. AND F. PINCHERA. 1991. Distribuzione e consistenza •AND -- 1996. Nest-site and aerial meeting delle popolazioni di alcune specie di Acdpitriformese point selection,nesting densityand reproduction of Falconiformesnidificanti nella regione Abruzzo. Pages Common Buzzrads.J. RaptorRes. 30:130-135. 363-366 in M. Fasola [ED.], Atti II Seminario Italiano CONE,C.D., JR. 1962. Thermal soaring in birds. Am. Sci. CensimentiFaunistici dei Vertebrati. Suppl. Ric. Biol. 50:180-209. Selvaggina,XVI: 1-820. GLUTZ VON BLOTZHEIM, U.N., K•M. BAUER AND E. BEZZEL. NEWTON,I., M. MARQUISS,D.N. WEIR AND D. MOSS. 1977 1971. Handbuch der Vtgel Mitteleuropas,Vol. 4. Spacing of Sparrowhawknesting territories.J. Atom. Frankfurt am Main, Germany. Ecol. 46:425-441. GOSZCZYNSKI,J. 1985. The effect of structural differen- --, P.E. DAVISANDJ.E. DAVIS. 1982. Ravensand buz- tiation of ecological landscapeon the predator-prey interactions. Publ. Warsaw Agric. Univ. SGGW-AR, zards in relation to sheep-farming and forestry in Warsaw (In Polish with English summary). Wales.J. Appl.Ecol. 19:681-706. G•,HAM, I.M., S.M. REDPATHAND S.J. THIRGOOD.1995. PANDOLFI,M. ANDG. TAFERNA.1991. Censimentodel po- The diet and breeding densityof Common Buzzards polamento di Passeriformidi un bosco di conifere e (Buteobuteo) in relation to indicesof prey abundance. di uno di latifogliedell'Italia centrale.Pages 375-378 Bird Study42:165-173. in M. Fasola [ED.], Atti II Seminario Italiano Cens•- HOHMANN,U. 1994. Status specific habitat use in the menti Faunistici dei Vertebrati. Suppl. Ric. Biol. Sel- Common Buzzard (Buteo buteo). Pages 359-366 in vaggina, XVI: 1-820. B.-U. Meyburg and R.D. Chancellor lEDS.I, Raptor PIcOZZl,N. ANDD. WEIR. 1974. Breeding biologyof the conservation today. Proc. IV World Conference on buzzardin Spaced.Breadboards 67:199-210. Birds of Prey and Owls. WWGBP/The Pica Press,Ber- ROBBINS, CS., D.K. DAWSON AND B.A. DOW•LL. 1989. lin, Germany. Habitat area requirementsof breeding forest birds of HUBERT,C. 1993. Nest-sitehabitat selectedby Common the middle Atlantic states.Wildl. Monog•.103:1-33. Buzzard (Buteobuteo) in southwesternFrance. J. Raptor ROCHg,J. 1977. Un recensementde busesen plaine de Res. 27:102-105. Sa6ne:quelques donn6es concernant la nidification. JANES,S.W. 1985. Habitat selection in raptorial birds. Jeande-Blanc16:48-63. Pages159-188 in M.L. Cody [ED.], Habitat selection SOKAL,R.R. AND FJ. ROHLF. 1981. Biometry, 2nd ed. in birds. Academic Press, London, UK. W.H. Freeman and Co., New York, NY U.S.A. JEDRZEJEWSKI,W., B. JEDRZEJEWSKAAND M. KELLER.1988. TAYLOR, K., R. HUDSON AND G. HORNE. 1988. Buzzard Nest-siteselection by the buzzard (Buteobuteo L.) in breeding distribution and abundance in Britain the extensive forests of eastern Poland. Biol. Conserv. and Northern Ireland in 1983. Bird Study 35:109- 43:145-158. 118. --, A. SZYMURAAND B. JEDREJEWSKA.1994. Repro- TITUS, K., J.A. MOSHER AND B.K. WILLIAMS. 1984. duction and food of the buzzard (Buteo buteo)in re- Chance-corrected classification for use in discrim•- 212 PE•4TEm•4• )a,•I) F•aYP,• VoI•. 31, No. 3

nant analysis:ecological applications. Amez.Midland WEIR, D. AND N. PICOZZI. 1983. Dispersion of buzzards Nat. 111:1-7. in Spaced.Brit. Birds76:66-78. Tu}•}•$,C.R. 1967. Population studyof buzzardsin the new forest during 1962-66. Brit. Birds60:381-395. 1974. The buzzard. David & Charles Limited, Received20 May 1996; accepted27 April 1997 Devon, UK.